GBT440, a Potent Anti-Sickling Hemoglobin Modifier Reduces Hemolysis, Improves Anemia and Nearly

Eliminates Sickle Cells in Peripheral Blood of Patients with Sickle Cell Disease

Claire Hemmaway, MA (Hons) Cantab, MBBS, FRCP, FRCPATH

Queens Hospital, Essex, United Kingdom

ASH 7 DECEMBER 2015

Josh Lehrer-Graiwer, MPhil, MD1, Jo Howard, MD2, Claire Hemmaway, MD3, Moji Awogbade, MD4, Paul Telfer, MD5, Mark Layton, MD6, Timothy Mant, MB7, Kobe Dufu, PhD1, Athiwat Hutchaleelaha, PhD1, Teri Koller1, Donna

Oksenberg, PhD1, Mira Patel, PhD1, Eleanor Ramos, MD1

1Global Blood Therapeutics, South San Francisco, CA, 2Guy’s Hospital, London, UK 3Queen’s Hospital, London, UK, 4Kings College Hospital, London, UK, 5Barts Hospital, London, UK, 6Hammersmith Hospital, London UK, 7Quintiles Drug Research Unit at Guy’s Hospital, London, UK

GBT440: Designed to Bind Hemoglobin with High Selectivity

2

• Reversible covalent binding to N-terminus of hemoglobin α chain stabilizes oxy-Hb conformation

• Profile confirms properties with high selectivity for hemoglobin

1:1 stoichiometry of GBT440 to Hb tetramer binding

Preferential partitioning into RBCs

Potent, dose-dependent increase in Hb-O2 affinity

GBT440 Clinical Hypothesis:Increase in HbO2 Affinity Inhibits HbS Polymerization

©2015 Global Blood Therapeutics3

Oxy-HbS monomer Deoxy-HbS polymer

HbSS RBC

Oxy-SS RBC Sickled RBC

Occluded Blood Flow

Hemolysis

GBT440 inhibits Hb polymerization

decreases RBC damage

Improvesblood flow

Stops hemolysis andimproves anemia

Modify course of SCD disease

GBT440-001: Study Design

• Randomized, double blind, placebo controlled study

• Population- Healthy subjects- SCD subjects

- HbSS, 18-60 yrs, Hb 6-10 g/dL, clinically stable (no VOC or RBC transfusion within 30 days), stable HU allowed

• Dose cohorts (8 subjects/cohort; 6 active:2 placebo)

• Objectives

• Safety

• Pharmacokinetics; pharmacodynamics (hemoglobin modification)

• SCD subjects: hematologic parameters

4

Demographics, Disposition and Efficacy data: unblinded by treatment assignment

Safety data: blinded evaluation

- Single dose cohorts Healthy subjects: 100, 400,

1000, 2000, 2800 mg SCD subjects: one cohort 1000

mg Subjects followed up to 28

days post dose

- Multiple dose cohorts Healthy subjects (15-day

dosing): 300, 600 and 900 mg SCD subjects (28-day dosing):

500, 700, 1000 mg Subjects followed up to 60

days post initiation of dosing

GBT440-001: Baseline Characteristics and Subject Disposition

Subject Disposition 500 mg 700 mg 1000 mg Placebo (pooled)

Dosed a 10 12 4 (ongoing) 9

Completed 28 day dosing 10 12 ongoing 8

Dose reduction 1b 1c 0 0

Completed 28 day follow-up 6 12 0 7

5

a No dose discontinuationsb One reduction from 500 mg to 200 mg due protocol specified reduction for asymptomatic Hb increase > 2 g/dLc One dose reduction from 700 mg to 400 mg due to abdominal cramps

Baseline Characteristics500 mg

n=10700 mg

n=121000 mg

n=4

Placebo (pooled)

n=9

Female n (%)Male n (%)

2 (20)

8 (80)

8 (67)

4 (33)

2 (50)

2 (50)

5 (56)

4 (44)

Age (years, median, range) 29 (19,47) 29 (20,56) 41 (36,45) 37 (20,53)

Weight (kg, median, range) 65 (53,75) 67 (53,83) 71 (68,85) 69 (57,88)

Hydroxyurea therapy n (%) 1 (10) 4 (33) 2 (50) 4 (45)

Baseline Hb (g/dL, median, range) 7.9 (7,9.7) 9.2 (7.5,9.8) 8.2 (7.5,8.4) 8.2 (7.2,10)

Hospitalizations due to painfulcrisis in prior year: n (%)01>2

7 (70)

3 (30)

0

5 (42)

3 (25)

4 (33)

4 (100)

0

0

4 (45)

3 (33)

2 (22)

Data as of 20 Nov 2015

GBT440-001: Safety Summary (blinded)

• Majority of adverse events were Grade 1 or 2 (mild-moderate) in severity

• Most common adverse event: headache

• 4 serious adverse events (all assessed not related to study drug)

• 3 sickle cell crisis events

1 unblinded (placebo treatment)

No sickle cell crisis have occurred during the 28-day treatment period

• 1 event of presumed infection with hemolysis

Unblinded (GBT440 treatment)

• No adverse effects on liver (LFTs) or renal (creatinine) function

• No evidence of tissue hypoxia

• No clinically significant ECG changes suggestive of ischemia

• No resting tachycardia, hypotension or increased respiratory rate

• No increase in erythropoietin or reticulocytes- these markers decrease with treatment suggesting improved oxygen delivery

6

Blinded data as of 20 Nov 2015

GBT440-001: Common Adverse Events (blinded)

Adverse events occurring >10% of subjects

7

Adverse Events (>10%)

500 mg GBT440 or placebo

n=14

700 mg GBT440 or placebo

n=16

All subjects(n=30)

Any AE 12 ( 85.7) 15 ( 93.8) 27 ( 90.0)

Headache 6 ( 42.9) 8 ( 50.0) 14 ( 46.6)

Back Pain 2 ( 14.3) 4 ( 25.0) 6 ( 20.0)

Pain* 2 ( 14.3) 4 ( 25.0) 6 ( 20.0)

Rhinitis 2 ( 14.3) 3 ( 18.8) 5 ( 16.7)

Fatigue 2 ( 14.3) 2 ( 12.5) 4 ( 13.3)

Sickle Cell Crises 1 ( 7.1) 3 ( 18.8) 4 ( 13.3)

Blinded data as of 20 Nov 2015

Sickle cell crises were Grade 3, all other adverse events were Grade 1 or 2*Pain events- majority assessed as sickle cell related

GBT440: Dose Proportional PK

• Dose proportional increase in GBT440 exposures following single and multiple dosing

• Half-life

• ~3 days healthy subjects

• ~1.6 days SCD subjects

• GBT440 RBC:plasma ratio ~75:1

8

GBT440 pharmacokinetics support once daily dosing

* Excludes subjects with dose reduction: 1 subject in 500 mg cohort and 1 subject in 700 mg cohort

0

50

100

150

200

250

300

350

400

500mg 700mg

Concnetration(µM)

GBT440WholeBloodConcentration

Mean(± SD)

500 mg (n=9)* 700 mg (n=11)*Cmax Cmin Cmax Cmin

GBT440 Whole Blood Concentrationat Steady State (mean ± SD)

Co

nce

ntr

atio

n (𝝁

M)

GBT440: Dose Proportional ‘Left Shift’ in P50

9

Representative Oxygen Equilibrium Curves

12.1mmHg

14.0mmHg

16.7mmHg

29.7mmHg

26.7mmHg

31.1mmHg

----GBT440 700 mg

----GBT440 500 mg

----Placebo

O2

Satu

rati

on

(%

)

pO2 (mm Hg)

Representative Oxygen Equilibrium Curves100

50

50 1000

p50 values

p20 values

• GBT440 results in left shift of the oxygen equilibrium curve SCD subjects are right shifted; p50 shifts to normal range (26-29 mm Hg)

• Hemoglobin modification is proportional to dose 500 mg (n=10): ~13% 700 mg (n=12): ~17%

p50 shifts to normal range

Inhibition of Hbs Polymerization Proof of Concept Biomarkers

10

Red blood cell damage

Hemoglobin andhemolytic markers

Sickle cells in peripheral blood

HbS

Reversibly sickled

Irreversibly sickled

Deoxygenated

HbS solution HbS polymers

Oxy-HbS monomer

Deoxy-HbS polymer

HbSS RBC

Oxy-SS RBC Sickled RBC

Occluded Blood Flow

Hemolysis

Data unblinded by treatment assignment1000 mg cohort enrollment in progress

GBT440 Treatment Increases Hemoglobin and Reduces ReticulocytosisConsistent with Reduction in Hemolysis

11

• Rapid rise in hemoglobin followed by decline in reticulocytosis is consistent with reduction in hemolysis

• Reduction in reticulocyte counts suggests improvement of red blood cell life span

Reduction in Hemolysis Correlates with GBT440 Exposure

12

Reticulocytes (%) vs GBT440 Exposure P

erce

nt

Ch

an

ge

in R

etic

ulo

cyte

s (%

) to

Da

y 2

8

60

40

20

0

-20

-40

-60

-80

0 25 50 75 100 125 150 175 200 225 250 275 300 325 350

GBT440 Whole Blood Concentration Cmin (𝜇M) at Day 28

PK data from 500 and 700 mg dose levels; R2 ~0.56

Higher GBT440 exposures resulted in more profound reduction in reticulocyte counts which is a measure of hemolysis and a surrogate of RBC lifespan

GBT440 Treatment Reduces Circulating Sickle Cells

13

Day -1

Representative images from GBT440 treated subject

Day 28 (GBT440 700 mg)

Circulating Sickle Cells (%)(relative change from baseline, median and 25th and 75th percentile )

D-1 D15 D28

DAYS

Per

cen

t C

ha

ng

e in

Sic

kle

Cel

ls (

%)

----GBT440 700 mg (n=12)

---- GBT440 500 mg (n=10)

---- Placebo (n=8)

Reduction in Unconjugated Bilirubin and LDH are Consistent with Reduced Intravascular Hemolysis

14

Reduction in bilirubin and LDH levels are observed as early as Day 4 and maintained throughout 28 day dosing period

Reduction in Erythropoietin is Consistent with Improved Oxygen Delivery to Tissues

15

Chan

ge in

Ery

thro

poie

tin -

(mU

/mL)

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

10

20

30

40

50

60

70

80

90

100

Days

-1 4 8 15 22 25 28

Absolute Change in Erythropoietin - (mU/mL), Median(with 25th and 75th percentiles)

Group: 700 mg 500 mg Placebo

Erythropoietin (mU/mL)(absolute change from baseline, median and 25th and 75th percentile)

---- GBT440 700 mg (n=12)

---- GBT440 500 mg (n=10)

---- Placebo (n=8)

D-1 D4 D8 D15 D28

Days

100

80

60

40

20

0

-20

-40

-60

-80

-100

Ab

solu

te c

ha

ng

e in

Ery

thro

po

ieti

n (

mU

/mL)

• Decline in Epo levels suggest that treatment with a Hb modifier that increases the affinity of Hb for oxygen leads to improvement of oxygen delivery in SCD subjects

• Reduction in Epo levels is sustained during the 28 day dosing period

Reduction in Soluble Adhesion Molecules Suggests Improvement in Inflammation

16

Per

cen

t C

han

ge

fro

m B

asel

ine

-30

-20

-10

0

10

20

30

40

50

60

70

Days

-1 15 28

Relative Change in P-Selectin, Median(with 25th and 75th percentiles)

Group: 700 mg Placebo

D-1 D15 D28

DAYS

P-Selectin (ng/mL)* (relative change from baseline, median and 25th and 75th percentile)

70

60

50

40

30

20

10

0

-10

-20

-30

Per

cen

t ch

an

ge

fro

m b

ase

line ---- GBT440 700 mg (n=12)

---- Placebo (n=4)

*Data currently available for subjects in 700 mg dose level

Per

cen

t C

han

ge

fro

m B

asel

ine

-15

-5

5

15

25

35

45

55

Days

-1 15 28

Relative Change in ICAM-1, Median(with 25th and 75th percentiles)

Group: 700 mg Placebo

ICAM-1 (ng/mL)*(relative change from baseline, median and 25th and 75th percentile)

---- GBT440 700 mg (n=12)

---- Placebo (n=4)

D-1 D15 D28

DAYS

55

45

35

25

15

5

-5

-15

Per

cen

t ch

an

ge

fro

m b

ase

line

• Reduction in soluble adhesion molecules (P-selectin, ICAM-1) may indicate decreased endothelial damage and inflammation

• Reduction in inflammation may lead to decreased risk of vaso-occlusion

Summary

• GBT440 was well tolerated over 28 days of dosing

• No drug-related serious adverse events

• No evidence of tissue hypoxia

• Decreases in erythropoietin and reticulocyte counts are consistent with improved oxygen delivery

• Pharmacokinetic data demonstrate linear and dose-proportional properties with a half-life amenable to once daily dosing

• Pharmacodynamic data (left shift in OEC) confirm pharmacological mechanism of action of GBT440

• GBT440 increases hemoglobin, reduces reticulocytosis and improves biomarkers of hemolysis and inflammation

• Hematologic effects are correlated with GBT440 blood levels

17

Conclusions

• Emerging data with 28 days of dosing support the hypothesis that GBT440 inhibits polymerization of sickle hemoglobin, improves hemolytic anemia, reduces red blood cell damage, and improves oxygen delivery

• Longer term dosing will determine the optimal magnitude of hematologic effects and clinical benefit

• GBT440 offers a well-tolerated, mechanism-based and potentially disease-modifying therapy for sickle cell disease

• Additional cohorts with longer term dosing planned

18

Acknowledgements

• Other referring hematologists

• Professor John Porter, University College London Hospitals

• Dr Marilyn Roberts-Harewood, North Middlesex University Hospital

• Other contributors• Global Blood Therapeutics: Brian Metcalf, Vincent Siu

• Quintiles, London: Cara Tite, Noel Landsman

• Professor Swee Lay Thein: King’s College Hospital (currently NIH/NHLBI)

• Professor David Rees: King’s College Hospital

• Subjects who participated in the study • Healthy volunteers

• Sickle cell patients

19